RIP ELEVENLABS! Create BEST TTS AI Voices LOCALLY For FREE!

Aitrepreneur
9 May 202417:45

TLDRThe video provides a comprehensive guide on how to create high-quality, custom text-to-speech (TTS) AI voices on your local computer for free. The host, SK, introduces various methods ranging from a quick 10-second voice cloning to a more sophisticated training of a personalized TTS model using only 2 minutes of audio. The video also covers the integration of the generated TTS audio with RVC (Reverse Voice Conversion) for enhanced voice quality. Additionally, it mentions an automatic method using the XTS RVC UI, which streamlines the process. The ultimate method, 'Uber text to speech,' combines fine-tuned TTS models with RVC for a highly authentic voice output. The guide is aimed at users looking for cost-effective alternatives to expensive AI voice services and offers a step-by-step approach to achieve professional-sounding voiceovers without the need for third-party software.

Takeaways

  • 📢 The video provides a tutorial on how to create custom text-to-speech (TTS) AI voices locally on your computer for free.
  • 🔧 Two installation methods are offered: a one-click installer for patrons and a manual installation process requiring Python, FFMpeg, and C++ build tools.
  • 🔊 A simple voice cloning method is demonstrated using just 10 seconds of audio clip through the XTTS web UI.
  • 🎓 The video also covers a medium TTS method where you can train your own XTTS model with only 2 minutes of audio.
  • ⚙️ For the ultimate TTS voice, the presenter outlines a combination of methods involving fine-tuning an XTTS model and using RVC (Reverse Voice Conversion).
  • 📈 The presenter emphasizes the importance of using a longer audio sample for better training results, recommending at least 10 minutes for higher quality.
  • 🌐 The video mentions the use of three different web UIs for different stages of the TTS creation process.
  • 🤖 An automated method using XTTS RVC UI is introduced, which simplifies the process by combining text-to-speech generation with voice conversion in one step.
  • 📝 The presenter provides tips on how to extend a short audio clip to the required duration for training by copying and pasting segments in Audacity.
  • 📚 A PDF guide will be available for free on the presenter's Patreon for those who need a visual reminder of the process.
  • 🌟 The ultimate goal is to achieve high-quality, authentic-sounding TTS voices without incurring high costs from third-party software.

Q & A

  • What is the main topic of the video?

    -The main topic of the video is about creating custom text-to-speech AI voices locally on your computer for free.

  • What are the different methods presented in the video for creating TTS AI voices?

    -The video presents several methods including a quick 10-second voice cloning, a medium method involving training your own XTTS model with 2 minutes of audio, and an ultimate method that combines fine-tuned XTTS models with RVC for high-quality voice cloning.

  • How much audio is required for the 'super lazy' method of voice cloning?

    -For the super lazy method, only 10 seconds of audio is required to clone a voice.

  • What software is mentioned for installing necessary components automatically?

    -FFMpeg and the C++ build tools are mentioned, along with a one-click installer for patrons and an ultimate text-to-speech auto installer.

  • How long does it take to generate a TTS voice using the quick cloning technique?

    -It takes only a few seconds to generate a TTS voice using the quick cloning technique, as demonstrated in the video.

  • What is the minimum duration of audio required to train your own XTTS model?

    -The minimum duration of audio required to train your own XTTS model is 2 minutes.

  • What is RVC and how is it used in the ultimate text-to-speech method?

    -RVC is a voice cloning software that can clone a voice to a near-perfect level. In the ultimate text-to-speech method, it is used to further refine the generated audio from the text-to-speech models to make it sound even more authentic.

  • How can you make the final audio file even better after generating it with a custom Obama model?

    -You can make the final audio file even better by downloading it and using RVC to select the reference voice model and convert the audio, resulting in a highly authentic and improved voice output.

  • What is the benefit of using the XTS RVC UI?

    -The XTS RVC UI automates the process of generating an XTTS audio and then converting it with RVC, making it easier and less time-consuming to produce high-quality voice-cloned audio.

  • How can you use the fine-tuned XTTS model in the XTS RVC UI?

    -You can use the fine-tuned XTTS model in the XTS RVC UI by copying the model files into the 'models xtts' folder within the XTS RVC UI directory, allowing the model to be used automatically within the interface.

  • What does the presenter suggest for those who want to remember the steps for creating TTS AI voices?

    -The presenter suggests that a PDF guide will be made available for free on their Patreon, which will help users remember the steps for creating TTS AI voices.

  • What is the presenter's recommendation for those who have questions or need support?

    -The presenter recommends that those with questions or needing support should send a direct message on Patreon, where they provide priority support to their patrons.

Outlines

00:00

🎙️ Custom Text-to-Speech AI Creation

The video begins by addressing the frustration with generic AI voices and the costs associated with them. It introduces the viewer to the possibility of creating custom text-to-speech AI voices using their local computer. The host, SK, promises to demonstrate various methods ranging from quick 10-second voice cloning to more sophisticated, high-quality voice generation. The paragraph outlines the preliminary step of installing necessary software, offering both a one-click installation for supporters and a manual installation process for others. It also guides on installing different Wave UIs and emphasizes the simplicity of the process.

05:02

🚀 Medium Text-to-Speech Method

This paragraph delves into the 'medium text-to-speech method', which involves training a custom text-to-speech model using just 2 minutes of audio. The process is made straightforward through the use of the xtts fine-tune web UI. The host shares a trick for those short on time, suggesting the duplication of a shorter audio clip to meet the 2-minute requirement. The training process is detailed, including the creation of a dataset, the training itself, and the optimization of the model. The outcome is a significantly improved voice clone that captures the nuances and characteristics of the original speaker's voice.

10:04

🎬 Advanced Text-to-Speech with RVC

The script moves on to the 'ultimate text-to-speech method', which combines text-to-speech generation with RVC (Reverse Voice Conversion) for enhanced voice cloning. It highlights three different methods within this approach, starting with a simple conversion using the xtts web UI and RVC. The process is further streamlined with the introduction of the XTS RVC UI, an automated tool that simplifies the conversion process. The paragraph concludes with the 'Uber text-to-speech method', which amalgamates all previous steps to achieve a highly refined text-to-speech model, using a fine-tuned xtts model and RVC for the final audio output.

15:06

📚 Conclusion and Additional Resources

The final paragraph wraps up the video with a summary of the methods presented for creating high-quality text-to-speech AI models on a local computer. It emphasizes the cost-effectiveness of these methods in comparison to third-party software subscriptions. The host offers a PDF guide for Patreon supporters to help remember the steps and encourages viewers to try out the methods for themselves. The video concludes with a call to action for viewers to subscribe, like, and support the channel for continued content creation.

Mindmap

Keywords

💡Text to Speech (TTS)

Text to Speech (TTS) is a technology that converts written text into audible speech. In the video, TTS is the central theme as the presenter discusses various methods to create high-quality AI voices for TTS using local computer resources without incurring high costs. The script mentions different levels of TTS, from quick cloning with a 10-second audio clip to training a personalized TTS model.

💡Voice Cloning

Voice cloning refers to the process of replicating a person's voice using AI and machine learning techniques. In the context of the video, the presenter demonstrates how to clone a voice with just 10 seconds of audio, which is a significant part of the discussed methods for creating custom TTS AI voices.

💡Local Computer

A local computer refers to a personal computer that is used on-site, not through remote access or the cloud. The video emphasizes the creation and utilization of TTS AI voices directly on one's local machine, thus avoiding the need for external services or online platforms.

💡XTTS (eXtreme TTS)

XTTS is a term used in the video to refer to an advanced text-to-speech system that the presenter uses to create and fine-tune AI voices. It is a core component in the methods described for generating custom TTS models on a local computer.

💡Voice Fine-Tuning

Voice fine-tuning is the process of training and adjusting a TTS model to better replicate a specific voice. The video script describes a method where one can fine-tune an XTTS model using just 2 minutes of audio to achieve a higher quality and more personalized voice output.

💡RVC (Resemblyzer Voice Cloning)

RVC, or Resemblyzer, is a voice conversion technology that allows for the creation of highly realistic voice clones. The video presents RVC as a tool to further improve the quality of the TTS AI voices by converting the generated audio to closely resemble a specific voice, such as that of Barack Obama.

💡One-Click Installer

A one-click installer is a software installation method that automates the process with a single user action, such as clicking a button. In the video, the presenter mentions the use of a one-click installer for supporters, simplifying the installation of necessary software for creating TTS AI voices.

💡Python

Python is a high-level programming language widely used for its simplicity and versatility. The script mentions the installation of Python as a prerequisite for manually setting up the environment to create TTS AI voices on a local computer.

💡FFMpeg

FFMpeg is a free and open-source project that deals with multimedia data, such as audio and video processing. It is mentioned in the script as a required software for the installation process to support the functionality of the TTS systems discussed in the video.

💡C++ Build Tools

C++ Build Tools are a set of compilers and libraries used for developing applications in the C++ programming language. The video script includes these tools as part of the software requirements for manually installing the necessary components for TTS voice creation.

💡Web UI

Web UI, or web user interface, refers to the interface of a web application that allows users to interact with the application through a web browser. The video demonstrates the use of Web UIs for the XTTS and RVC systems to provide a user-friendly way to create and manipulate TTS AI voices.

Highlights

Create custom text-to-speech AI voices on your local computer for free.

Explore a range of methods from quick 10-second voice cloning to the ultimate text-to-speech voice.

Install necessary software using one-click installer for Patreon supporters or manual installation.

Use FFMpeg and C++ build tools for the manual installation process.

Clone a voice with just 10 seconds of audio using the XTTS web UI.

No character limit for text input in the simple text-to-voice tab.

Train your own text-to-speech model with only 2 minutes of audio using the XTTS fine-tune web UI.

Use Audacity to extend a shorter audio clip into the required 2-minute training length.

Fine-tuning captures the speaker's accent, speech patterns, and unique quirks.

Convert generated text-to-speech audio to a desired voice using RVC (Resemblyzer Voice Converter).

Automatically generate and convert audio with the XTTS-RVC UI for a streamlined process.

Combine fine-tuned XTTS models with RVC for the ultimate text-to-speech combination.

Use the same fine-tuned model for multiple projects without limitations.

The final Uber text-to-speech method provides the highest level of quality and authenticity.

No need to pay exorbitant fees for third-party software with these local text-to-speech solutions.

A PDF guide will be available for free on Patreon for easy reference.

Patreon supporters get priority support and access to additional resources.

The video provides a comprehensive guide to creating high-quality AI voices without expensive software.